researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PerCom53586.2022.9762376

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sensors  

Information on congestion of buses, which are one of the major public transportation modes, can be very useful in light of the current COVID-19 pandemic. Because it is unrealistic to manually monitor the number of riders on all buses in operation, a system that can automatically monitor congestion is necessary. The main goal of this paper’s work is to automatically estimate the congestion level on a bus route with acceptable performance. For practical operation, it is necessary to design a system that does not infringe on the privacy of passengers and ensures the safety of passengers and the installation sites. In this paper, we propose a congestion estimation system that protects passengers’ privacy and reduces the installation cost by using Bluetooth low-energy (BLE) signals as sensing data. The proposed system consists of (1) a sensing mechanism that acquires BLE signals emitted from passengers’ mobile terminals in the bus and (2) a mechanism that estimates the degree of congestion in the bus from the data obtained by the sensing mechanism. To evaluate the effectiveness of the proposed system, we conducted a data collection experiment on an actual bus route in cooperation with Nara Kotsu Co., Ltd. The results showed that the proposed system could estimate the number of passengers with a mean absolute error of 2.49 passengers (error rate of 38.8%).

DOI： 10.3390/s22030881

Web of Science

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3460418.3479280

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3472306.3478355

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.3390/s21175719

出版者・発行元：Multimodal Technologies and Interaction  

Subtitles play a crucial role in facilitating the understanding of visual content when watching films and television programs. In this study, we propose a method for presenting subtitles in a way that considers cognitive load when viewing video content in a non-native language. Subtitles are generally displayed at the bottom of the screen, which causes frequent eye focus switching between subtitles and video, increasing the cognitive load. In our proposed method, we focused on the position, display time, and amount of information contained in the subtitles to reduce the cognitive load and to avoid disturbing the viewer’s concentration. We conducted two experiments to investigate the effects of our proposed subtitle method on gaze distribution, comprehension, and cognitive load during English-language video viewing. Twelve non-native English-speaking subjects participated in the first experiment. The results show that participants’ gazes were more focused around the center of the screen when using our proposed subtitles compared to regular subtitles. Comprehension levels recorded using LightSub were similar, but slightly inferior to those recorded using regular subtitles. However, it was confirmed that most of the participants were viewing the video with a higher cognitive load using the proposed subtitle method. In the second experiment, we investigated subtitles considering connected speech form in English with 18 non-native English speakers. The results revealed that the proposed method, considering connected speech form, demonstrated an improvement in cognitive load during video viewing but it remained higher than that of regular subtitles.

DOI： 10.3390/mti8060051

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1109/JIOT.2023.3347552

出版者・発行元：2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2024  

In recent years, with the trend towards shorter working hours, the quality of meetings has increasingly impacted these hours. Consequently, meeting quality and value become more important. To achieve better results in limited time, everyone needs to conduct meetings effectively. This requires skills such as facilitating the meeting, listening to others' opinions, accurately expressing one's own views, and using body language. However, these skills are currently only correlated with meeting effectiveness and lack comprehensive qualitative and quantitative assessments, leaving specific methods for improvement unclear. Additionally, the type of meeting can lead to participant stress or disinterest, making it essential to understand their psychological safety and engagement for more effective meetings. Measuring these psychological states poses significant challenges due to privacy and compliance concerns, particularly when using cameras or wearable sensors. This paper broadly addresses these issues and reports on our approach to detecting non-verbal cues, specifically nodding, from chair movements as a potential solution.

DOI： 10.1109/PerComWorkshops59983.2024.10502817

Web of Science

Scopus

掲載種別：研究論文（学術雑誌）   出版者・発行元：IEEE Access  

The development of smart sensors and appliances can provide a lot of services. Nevertheless, the act of aggregating data containing sensitive information related to privacy in a single location poses significant issues. Such information can be misused by a malicious attacker. Also, some previous studies attempted to apply privacy mechanisms, but they decreased data utility. In this paper, we propose privacy protection mechanisms to preserve privacy-sensitive sensor data generated in a smart home. We leverage Rényi differential privacy (RDP) to preserve privacy. However, the preliminary result showed that using only RDP still significantly decreases the utility of data. Thus, a novel scheme called feature merging anonymization (FMA) is proposed to preserve privacy while maintaining data utility by merging feature dataframes of the same activities from other homes. Also, the expected trade-off is defined so that data utility should be greater than the privacy preserved. To evaluate the proposed techniques, we define privacy preservation and data utility as inverse accuracy of person identification (PI) and accuracy of activity recognition (AR), respectively. We trained the AR and PI models for two cases with and without FMA, using 2 smart-home open datasets i.e. the HIS and Toyota dataset. As a result, we could lower the accuracy of PI in the HIS and Toyota dataset to 73.85% and 41.18% with FMA respectively compared to 100% without FMA, while maintaining the accuracy of AR at 94.62% and 87.3% with FMA compared to 98.58% and 89.28% without FMA in the HIS and Toyota dataset, respectively. Another experiment was conducted to explore the feasibility of implementing FMA in a local server by partially merging frames of the original activity with frames of other activities at different merging ratios. The results show that the local server can still satisfy the expected trade-off at some ratios.

DOI： 10.1109/ACCESS.2024.3390039

Web of Science

Scopus

researchmap

出版者・発行元：MOBISYS 2024 - Proceedings of the 2024 22nd Annual International Conference on Mobile Systems, Applications and Services  

Since office work has become large-scale and diversified in companies or organizations, work engagement and efficiency have been always an important index of a team's or group's evaluation because it is directly connected to their outcomes. In order to identify the group work context, we first need to recognize for what and how long the individual members are spending their time at their desks, but without privacy concerns and underestimation of their actual work. In this paper, we propose and evaluate the base system of personal desk activity recognition by using a low-cost compact WiFi node and its WiFi channel state information (CSI), which can lead to a lightweight group work context identification system. As a result, we achieved 94.2% desk activity recognition accuracy using the on-desk receiver, in recognizing five different classes.

DOI： 10.1145/3643832.3661452

Web of Science

Scopus

出版者・発行元：IEEE Access  

Machine learning's exponential growth has transformed healthcare, with Federated Learning (FL) playing a pivotal role. Despite its significance, FL is vulnerable to privacy attacks. In response, researchers have integrated differential privacy (DP) into FL. Nevertheless, incorporating DP introduces challenges such as increased total communication costs and computational overheads due to the introduction of noise. This drawback renders FL with DP less viable for healthcare systems, characterized by numerous low-resource devices and network bandwidth constraints. To overcome this limitation, we propose integrating a Discrete Fourier Transform (DFT) aggregator post-noise addition to transform the gradient generated by local training before sending it to the central server. This process reduces the gradient size and provides rudimentary encryption. The evaluation results reveal the superior performance of our proposed method, demonstrating an enhanced accuracy ranging from 0.2% to 2% compared to existing differential privacy techniques, including RDP, DP-SGD, ZcDP, LDP-Fed, and DP-AdapClip. Our approach substantially reduces the total communication costs (ranging from 6% to 43% across different privacy budgets) with faster training times in healthcare datasets such as the PIMA Indian database and Breast Cancer Histopathology Images.

DOI： 10.1109/ACCESS.2024.3418016

Web of Science

Scopus

出版者・発行元：International Conference on Information Networking  

As the high-speed wireless network spreads, the usage of shared Wi-Fi in places such as offices, coworking spaces, and homes has been increased. Along with the development and diversification of digital content, the amount of data consumed by individual devices has also grown. In situations where multiple users share limited network resources, some content that consumes a large amount of bandwidth like video streaming can potentially degrade the Quality of Experience (QoE) for other users nearby. Consequently, a need for persuasive intervention systems that encourage prosocial behavior is rising taking into consideration the QoE of other network users. To address this issue, this study proposes a "Persuasive Public Wi-Fi"that employs Wi-Fi to incrementally convince users towards more considerate usage of network resources. To be specific, we suppose that the persuasive public Wi-Fi must include three different modes: 1) a mode of normal networking, 2) a mode that can intervene with individual users through a captive portal, and 3) a mode that intentionally limits bandwidth, i.e., Quality of Service (QoS) control. This work demonstrates the design and implementation of the proposed system and presents the results of operational verification using a prototype system.

DOI： 10.1109/ICOIN59985.2024.10572181

Scopus

出版者・発行元：MOBISYS 2024 - Proceedings of the 2024 22nd Annual International Conference on Mobile Systems, Applications and Services  

We present a secure decentralized learning framework suitable for resource-constrained devices within a cluster environment. Our approach focuses on enhancing privacy preservation during model aggregation by utilizing Differential Privacy. This technique adds random noise to gradients obtained from local training on edge devices before sending them for aggregation. This noise addition ensures that sensitive information within the gradients remains distorted, thus safeguarding user privacy. We showcase the implementation of our system on a cluster system employing Raspberry Pi 4 Model B devices, illustrating its feasibility and effectiveness in real-world scenarios. Through this demonstration, we highlight the practical applicability of our system in enabling secure decentralized learning within resource-constrained environments.

DOI： 10.1145/3643832.3661843

Web of Science

Scopus

出版者・発行元：CEUR Workshop Proceedings  

In an era where digital technology profoundly influences our daily habits, we explore the potential of color psychology as a key element in the design of persuasive technology to promote health behavior change. This paper presents a strategy that uses color psychology to both encourage positive behaviors and discourage negative ones. We present two different applications: "eat2pic," a system designed to promote healthier eating habits by transforming meal selection into an engaging activity of adding color to images, and "color-wall," an application designed to reduce digital distractions by applying a grayscale filter to non-essential digital content. Through these demonstrations, we explore the efficacy and potential of color psychology in persuasive technology to influence user behavior in a health context and offer promising solutions to improve dietary choices and concentration while using computers.

Scopus

掲載種別：研究論文（学術雑誌）   出版者・発行元：IEEE Access  

Sugary beverages are a significant contributor to sugar consumption, and their excessive consumption is associated with increased risks of elevated blood glucose levels and diabetes. Many individuals have a strong preference for sugary beverages and often find beverages with lower sugar content to be less satisfying. Attempts to switch to less sugary options are frequently short-lived, leading to a return to higher-sugar beverages. Recognizing that 75 - 95% of taste perception is influenced by scent, we investigated a scent-based approach to reduce sugar intake while preserving the perception of sweetness. This study introduces an olfactory interface in the form of a mug named 'Aromug,' designed to emit a sweet scent in sync with the drinking action. Aromug incorporates motion sensing and scent presentation functions to enhance the perceived sweetness of a beverage, thereby encouraging a gradual reduction in sugar intake. Our experiments, involving 33 participants, demonstrated that the combined scents of sugar-free coffee and chocolate increased the perception of sweetness (p = 1.641 × 10-2). The study also found that the simultaneous presentation of scent and visual cues improved taste satisfaction and sweetness perception. Additionally, we observed variations in sweetness preference related to age and frequency of coffee consumption. It was particularly observed that people in their 20s and those who frequently drink coffee tend to perceive the taste of beverages as sweeter. This suggests a potential for Aromug to customize the scent experience based on individual preferences, offering a novel way to encourage healthier beverage choices.

DOI： 10.1109/ACCESS.2024.3401392

Web of Science

Scopus

researchmap

出版者・発行元：2024 8th International Conference on Robotics, Control and Automation, ICRCA 2024  

Dense video captioning tasks are used to detect interesting events and provide descriptive text for these events from untrimmed videos. This technology has the potential to be used in security surveillance and human care applications. However, current methods often overlook the relationships between objects in the video, which limits their applicability and makes it challenging to adapt them to specific domains, such as video summarization for indoor human activities. In these scenarios, human activities are closely intertwined with the objects in the scene. In this paper, we propose a plug-and-play module designed to enhance existing dense video captioning methods with spatial information. Specifically, we extract spatial information about the interesting objects using Red-Green-Blue-Depth (RGB-D) images and the results of image segmentation. We then integrate this information into the captions generated by the Dense Video Captioning (DVC) method using a fine-tuned Large Language Model (LLM). We evaluate the performance of our model on a custom dataset and demonstrate that our system provides a convenient and effective approach for obtaining space-enhanced captions.

DOI： 10.1109/ICRCA60878.2024.10649311

Scopus

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM International Conference Proceeding Series  

In this paper, we introduce the concept of a spoken dialogue system, which aims to recognize user eating behaviors during meals and promotes healthier eating habits by engaging users in an interactive co-dining experience. To provide an intuitive and more natural access, this information is integrated into a natural conversation with the user. Therefore, we conducted a preliminary study to explore how a human-like virtual avatar's eating speed affects the eating pace of a co-dining user. The results indicate that dining together with an avatar reduces the user's feeling of loneliness and enhances the overall satisfaction and enjoyment of the meal. Therefore, this work represents an important step toward realizing engaging co-dining experiences with human-like avatars and provides new insights into the future design of IoT systems to support solitary dining and health recommendations.

DOI： 10.1145/3627050.3630728

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/iot/iot2023.html#MayumiN0AYM23

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM International Conference Proceeding Series  

To support the decision-making process for selecting upcoming tourist spots (PoI: Points of Interest), it is necessary to share the tourist destinations' current status. Methods for sharing the current status of PoIs could include installing live cameras or having tourists share real-Time photos. Still, there are challenges in terms of installation and maintenance costs and privacy concerns. In this paper, a tourism support system is proposed to aid in the decision-making of which PoI to visit next by automatically generating PoI current status images from template images and contextual information of tourist destinations while preserving privacy and presenting them to users. In the proposed system, semantic segmentation is used to divide template images into categories such as sky class, trees/grasses class, crowd class, and overall class. Based on the separately collected contextual information (weather, condition of trees/grass, crowd density, time, etc.), appropriate image transformations (using style transfer, etc) are applied to each category. The proposed system consists of an automated method for generating PoI status images and an application that presents these images to users. As part of a tourism experiment using the proposed system, 11 individuals aged 20 to 30 participated in a roughly 2-hour tour of Nara City in Japan. The results showed that compared to presenting template images, the proposed method yielded a decision-making score improvement of 0.27 and a similarity (to the current status) score improvement of 0.45 (out of five), both of which were statistically significant.

DOI： 10.1145/3627050.3627068

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/iot/iot2023.html#KawanakaNSY23

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：ACM International Conference Proceeding Series  

In this demo, we will show our proposed system that can count nodding without either a camera or any sensor attached to the person. Our proposed system capitalizes on the fact that the upper body moves in conjunction with nodding and that this body motion slightly shakes the chair. We explore the challenge of recognizing nodding from the extremely subtle sway of a chair. To recognize nods in real-Time, we employed a supervised learning approach using acceleration data from sensors attached to the chair's backrest. Ultimately, the Support Vector Machine (SVM) achieved a nodding recognition accuracy of 0.990. Further testing of the accuracy of nodding frequency measurements yielded an accuracy of 0.947, suggesting that the optimal position for the accelerometer is the backrest. These results indicate that simply placing the accelerometer on the backrest can effectively quantify the frequency of nods from seated participants.

DOI： 10.1145/3627050.3630740

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/iot/iot2023.html#HayashidaNCA23

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sustainability (Switzerland)  

Litter has become a social problem. To prevent litter, we consider urban planning, the efficient placement of garbage bins, and interventions with litterers. In order to carry out these actions, we need to comprehensively grasp the types and locations of litter in advance. However, with the existing methods, collecting the types and locations of litter is very costly and has low privacy. In this research, we have proposed the conceptual design to estimate the types and locations of litter using only the sensor data from a smartwatch worn by the user. This system can record the types and locations of litter only when a user raps on the litter and picks it up. Also, we have constructed a sound recognition model to estimate the types of litter by using sound sensor data, and we have carried out experiments. We have confirmed that the model built with other people’s data enabled to estimate the F-measure of 80.2% in a noisy environment through the experiment with 12 participants.

DOI： 10.3390/su151310079

Web of Science

Scopus

researchmap

Web of Science

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 14th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2023  

Internet use is expanding along with the spread of smartphones and the promotion of digital use, but information security literacy is required to use the Internet safely. This paper addresses these challenges by focusing on the potential of microlearning and nudge theory to support continuous learning to solve these problems and presents a novel learning system that utilizes a Wi-Fi landing portal that appears as a dialog or a browser. To assess the effectiveness of the proposed system, we conducted a comparative analysis of information security test scores before and after the operation of the learning system and questionnaire-based evaluation. The results showed that the analysis of answer scores before and after the operation of the learning system was significant (\mathrm{p} < 0.05), indicating an improvement in scores. It was also found that using interface information with nudges effectively motivated participants to engage in repeated learning behaviors. However, it should be noted that frequent interventions might reduce acceptability, and the psychological burden arising from task interruptions needs to be taken into account.

DOI： 10.23919/ICMU58504.2023.10412210

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icmu/icmu2023.html#HiraokaMAN23

Web of Science

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 14th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2023  

Nowadays more and more people are wearing smart-watches in their daily lives. The various sensors embedded in smartwatches bring the ability to evaluate users' status as well as the risk of privacy issues. For example, if users are typing on key-boards while wearing smartwatches, the attacker could know the typing contents from the sensor data collected by the malicious applications that are installed on the targets' smartwatches. In this paper, we propose WatchLogger, the framework using audio and accelerometer signals to recognize the English words being typed, for demonstrating how to implement the smartwatch-based side-channel attack. Different from the previous studies that focused on the recognition of each key or pair of keys being pressed, WatchLogger aims to perform recognition on the scale of words. To achieve this goal, WatchLogger exploits the audio signals for segmentation and the accelerometer signals for classification. In addition, we propose an ensemble classification model to deal with the problem caused by too many words. At last, we build the dataset WTW-100 with 100 classes of words and 100 samples for each class, and we conduct experiments on the dataset. The experimental results show an accuracy of 98.5 % for keystroke recognition and 91.5 % for word classification, showing a considerable performance of WatchLogger.

DOI： 10.23919/ICMU58504.2023.10412218

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icmu/icmu2023.html#LiANCFW23

Web of Science

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Proceedings - International Conference on Tools with Artificial Intelligence, ICTAI  

In this paper, we propose ToonMeet, a hybrid frame-work for high-resolution and style-controllable online meeting toonification that ensures real-time operation speed. ToonMeet applies video frame interpolation to traditional portrait toonification pipelines, allowing for the synthesis of intermediate frames between adjacent toonified keyframes, significantly accelerating the overall process and saving computational resources. However, this approach brings a new problem, where prevailing flow-based video frame interpolation methods tend to cause more ghost and blur artifacts in toonified scenes compared to non-toonified scenes, especially when fast-moving objects exist. We study this previously undiscussed problem and explore its causes. To address this, we introduce a new dataset called TM3B (Toonified Multi-modal Meeting Behaviors), offering high-resolution and cross-platform multi-modal stylized meeting data of Japanese youth in various scenarios. Then, we fine-tune ToonMeet on these tailored data and the resulting model presents improved optical flow estimation ability on toonified videos. Extensive experiments demonstrate that ToonMeet can achieve great spatiotemporal performance and perform high-quality toonification of online meetings with real-time operation speed.

DOI： 10.1109/ICTAI59109.2023.00013

Web of Science

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/ictai/ictai2023.html#ChenFNA23

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

In this paper, we propose a purchasing support system for offline stores that can present information to users without being evasive by interactively presenting a small amount of product information according to their interest in the product as predicted by the system. We design and implement a proof-of-concept system for an actual use case scenario: a snack purchasing area in an office. We evaluated the effectiveness of the proposed system through a snack purchasing experiment with 11 participants. Experimental results suggest that the proposed method can induce user interest and encourage viewing and that some users may receive and consider the information presented while making a purchase.

DOI： 10.1007/978-3-031-30933-5_8

Web of Science

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/persuasive/persuasive2023.html#TokunagaMSOYNA23

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023  

3D applications such as VR and AR are attracting increasing commercial attention, and point cloud video is expected to be one of the most suitable representations for real-time applications due to its simplicity and versatility. However, point cloud data is large in size and difficult to stream in a mobile network environment with limited bandwidth. Therefore, a method for streaming point clouds with low bandwidth consumption while maintaining the quality of the user experience is needed. In this paper, we present a point cloud streaming method of real-space objects such as humans and animals for real-time 3D reconstruction in VR space. The system uses a depth camera to scan a human or animal, divides the point cloud into parts of the body, and then controls the quality of the point cloud (i.e., resolution and frame rate) for each part in real-time according to the object's motion and context. This enables point cloud streaming with limited resources (computational and network resources) and maximizes the user's quality of experience. We exhibit a series of systems that enhance the user experience in remote communication in realistic environments and scenarios while maintaining interactivity between real-space objects and remote users.

DOI： 10.1109/PerComWorkshops56833.2023.10150405

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/percom/percomw2023.html#IshimaruNFSY23a

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023  

There are two nasal pathways through which we perceive odors: 'orthonasal smell,' which comes from the tip of the nose, and 'retronasal smell,' which comes from the mouth and exits through the nose. In particular, the retronasal smell is known to be more critical in influencing perceived taste than the orthonasal smell. In this study, we propose a smart lid called Kaolid that changes the taste of beverages by presenting scent information as retronasal smell. Kaolid achieves retronasal smell by equipping a small, lightweight straw- or cup-type olfactory device with a sensor that senses drinking behavior and the amount of water in the cup. The straw-type device injects the scent into the tube through the middle of the straw, delivering the scent to the user's mouth simultaneously as the beverage is being consumed. The cup-type device injects the scent into the cup through the lid and delivers the scent to the user's mouth at the beginning of drinking to achieve in-mouth scenting. The concept of Kaolid is to promote water consumption, considered insipid, by altering the sense of taste through a scent presented when drinking the beverage.

DOI： 10.1109/PerComWorkshops56833.2023.10150412

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/percom/percomw2023.html#MayumiNMUMY23

掲載種別：研究論文（国際会議プロシーディングス）  

Federated learning has gained popularity as a distributed machine learning approach that provides security and privacy for data trained on local devices. However, vulnerabilities still exist in this approach, and common solutions such as encryption and blockchain techniques often suffer from high computation and communication costs, making them impractical for resource-constrained devices. To solve this problem, we propose a privacy-preserving federated learning system that leverages compressive sensing and differential privacy, specifically designed for devices with limited computational resources. In this paper, we demonstrate the capabilities of our proposed system in resource-limited environments. We outline the features, infrastructure, and algorithm of our proposed system, and simulate its performance using image datasets on a Raspberry Pi 4 and an Android smartphone in a cloud environment. Our approach offers a practical solution for secure and privacy-preserving federated learning in resource-constrained scenarios, with potential applications in various domains such as healthcare, IoT, and edge computing.

DOI： 10.1109/MDM58254.2023.00038

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/mdm/mdm2023.html#HidayatNA23

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023  

Many services based on human activity recognition (HAR) have been developed; however, user activity data include a large amount of private information. Although privacy protection is important in activity recognition, it has not been sufficiently explored. Therefore, we propose a privacy-preserving mechanism for HAR services that uses differential privacy. The proposed method reduces the user recognition accuracy to a level that satisfies the privacy requirements by adding weighted noise to the features in the learning model construction and then improves the activity recognition accuracy (service usefulness). The results indicate that when the privacy requirement is defined as less than the probability of a user being identified by chance, the proposed method improves the activity recognition accuracy by approximately 10 % compared to the conventional method.

DOI： 10.1109/PerComWorkshops56833.2023.10150239

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/percom/percomw2023.html#FujimotoNA23

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 14th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2023  

In this paper, we propose a method to measure human head motion, especially nodding, without attaching any sensors to the person. Our proposed system focuses on the fact that the upper body moves along with nodding and that the body motion slightly shakes the chair. We challenge the problem of whether it is possible to recognize a nodding from the extremely slight sway of a chair. To reveal the optimal position of sensors, we collected data by attaching multiple accelerometers to various positions on a chair, including the backrest, the seat's underside, and the legs. Using a supervised learning approach, we determined the best positions and combinations of sensors for recognizing nodding more collectively. The Support Vector Machine (SVM) achieved a nodding recognition accuracy of 0.990. Further testing of the accuracy of nodding frequency measurements resulted in an accuracy of 0.947, suggesting that the best position for the accelerometer is the backrest. These results suggest that simply placing the accelerometer on the backrest can effectively quantify the nod frequency of seated participants.

DOI： 10.23919/ICMU58504.2023.10412249

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/icmu/icmu2023.html#HayashidaNCA23

Web of Science

出版者・発行元：CEUR Workshop Proceedings  

In this study, the target behavior of prosocial behavior was defined as garbage disposal (behavior of taking out the trash), and an intervention method using an inducement was proposed and evaluated. Specifically, we conducted an intervention in which the saturation of a trash can was visualized using a colored light that could be changed to red, yellow, and green. In addition, questionnaire surveys were administered before and after the intervention to 28 undergraduate and graduate students and faculty members, asking about their attitudes and feelings toward garbage disposal behavior, and personality traits. The results of a survey on the frequency of trash disposal be-fore and after the experiment showed that the time that the saturated trash can was left unattended was reduced by 81%. The questionnaire survey confirmed that the intervention increased the positive attitude toward trash disposal behavior, indicating that the intervention had an impact on trash disposal behavior. In addition, we were able to confirm the characteristics of individual personality traits and susceptibility to intervention.

Scopus

出版者・発行元：Proceedings - IEEE International Conference on Mobile Data Management  

Federated learning provides techniques for training algorithms using mobile or decentralized devices, in contrast to traditional machine learning in which algorithm training is performed on centralized devices. In addition, federated learning provides privacy and security features, as the client and server do not share raw data, which may contain confidential information. A number of studies have shown, however, that using federated learning alone is not enough to protect data privacy in certain situations. To overcome this problem, differential privacy is proposed, which is a technique in which artificial noise is added to the raw data. By implementing this method, a high level of privacy protection can be obtained, however this added noise also reduces model accuracy. To address this issue, this paper proposes a new approach to implement differential privacy in federated learning using adaptive Gaussian clipping. We implemented the method by tightening the privacy budget, and introducing dynamic sampling probability, adaptive clipping based on hyperparameters, and a new privacy loss calculation. Our method's main objective is to adaptively change the amount of noise given to the model, thereby maximizing the model's accuracy performance, while maintaining privacy protection levels. Evaluation results show that our proposed method presents slightly better accuracy when compared to other existing differential privacy variants such as RDP, DP-SGD, and ZcDP, for both balanced (i.i.d.) and unbalanced datasets (non-i.i.d.), for a lower total communication cost than some variants.

DOI： 10.1109/MDM58254.2023.00042

Scopus

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2023  

In this paper, we present a point cloud streaming method of real-space objects such as humans and animals for real-time 3D reconstruction in VR space. The system uses a depth camera to scan a human or animal, divides the point cloud into parts of the body, and then controls the quality of the point cloud (i.e., resolution and frame rate) for each part in real-time according to the object's motion and context. This enables point cloud streaming with limited resources (computational and network resources) and maximizes the user's quality of experience (QoE). We have implemented and evaluated a series of systems incorporating the proposed method to enhance the user experience in realistic environments and scenarios while maintaining interactivity in VR-based online communication. The results show that the proposed system is feasible under resource-constrained environments without significantly affecting the user's QoE.

DOI： 10.1109/PerComWorkshops56833.2023.10150337

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/percom/percomw2023.html#IshimaruNFSY23

記述言語：英語  

DOI： 10.1007/978-3-030-94822-1_50

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：UbiComp/ISWC 2022 Adjunct - Proceedings of the 2022 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2022 ACM International Symposium on Wearable Computers  

As the Attention Economy has grown, numerous digital contents are made with the intention of capturing users' attention and spending more of their time by the use of attractive graphics and color psychology. In this paper, we propose a novel defense strategy, called a color-wall, to block digital distraction brought on by the consumption of visually seductive and addictive digital content. When it detects that a user is viewing media or using applications unrelated to the main task, it removes colors from the screen (gray-scaling) to make them less aesthetically pleasing and directs the user away from looking at them. We implemented a mac OS software as a proof-of-concept for the color-wall and conducted user testing (N=14) to assess the effectiveness of the proposed strategy. The evaluation results suggest that the color-wall is an effective strategy for reducing unwanted digital distractions such as browsing restricted content (YouTube, Twitter, etc.) while working on a PC.

DOI： 10.1145/3544793.3563403

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/huc/ubicomp2022ap.html#NakamuraTA22

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：UbiComp/ISWC 2022 Adjunct - Proceedings of the 2022 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2022 ACM International Symposium on Wearable Computers  

Reducing the amount of sugar consumed from daily beverages is essential for well-being. However, since a sudden reduction in intake is mentally painful for those who prefer sweetened beverages, solutions are needed to help them gradually become accustomed to non-sweetened drinks. In this paper, we propose Aromug, a mug-type olfactory interface that amplifies the user's perceived sweetness by providing a sweet scent along with the action of drinking. Through user studies, we investigated the effect of the information presented by Aromug on the taste of iced coffee. The experiment results showed that the sugar-free coffee and chocolate aroma might enhance the perceived sweetness compared to the no-scent case. In addition, it was suggested that there were differences in preference for sweetness depending on age and frequency of coffee drinking.

DOI： 10.1145/3544793.3563402

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/huc/ubicomp2022ap.html#MayumiN0MY22

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sensors and Materials  

Short stick exercises have been attracting attention from the viewpoint of preventing falls and improving the health of elderly people and are generally performed under the guidance of instructors and nursing staff at nursing homes. However, in situations such as the COVID-19 pandemic, where people should refrain from unnecessary outings, it is advisable that individuals perform short stick exercises at home and record their exercise implementation status. In this paper, we propose an inertial measurement unit (IMU)-sensor-based short stick exercise tracking method that can automatically record the types and amounts of exercises performed using a short stick equipped with an IMU sensor. The proposed method extracts time-domain and frequency-domain features from linear acceleration and quaternion time-series data obtained from the IMU sensor and classifies the type of exercise using an inference model based on machine learning algorithms. To evaluate the proposed method, we collected sensor data from 21 young subjects (in their 20s) and 14 elderly subjects (79-95 years old), where the participants performed three sets (10 times per set) of eight basic types of short stick exercises (five types for elderly people). As a result of evaluating the proposed method using this data set, we confirmed that when LightGBM was used as the learning algorithm, it achieved F values of 90.0 and 86.6&#37; for recognizing the type of exercise for young and elderly people, respectively.

DOI： 10.18494/sam3968

Web of Science

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Proceedings - 2nd International Workshop on Cyber-Physical-Human System Design and Implementation, CPHS 2022  

Stick exercises, which have been attracting attention for improving the health of the elderly, are usually performed in nursing homes under the guidance of nursing staff. However, in the current pandemic in which the elderly are advised to refrain from going out unnecessarily, it is desirable for each individual to be able to perform the stick exercises alone. In this study, we aim to develop a stick exercise support system that can automatically record the number of times an elderly person performs each type of stick exercise and provide feedback to improve the movement for each exercise. As a first step toward the realization of this stick exercise support system, we investigated a method for recognizing exercise movements using inertial measurement unit (IMU) sensors. In the evaluation experiment, 21 subjects performed 3 sets (10 times per set) of eight basic stick exercises. The exercise movements were classified based on the linear acceleration and quaternion data obtained from the IMU. As a result, 90&#37; of F-measure was achieved when using Light GBM as the learning algorithm.

DOI： 10.1109/cphs56133.2022.9804564

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Proceedings - 2nd Workshop on Data-Driven and Intelligent Cyber-Physical Systems for Smart Cities Workshop, DI-CPS 2022  

In many of the world's major cities, commuter trains provide vital transportation support and thus play an essential role in our daily lives. Therefore, it has become necessary to estimate the degree of congestion in each train car, both to improve passenger comfort levels and, more recently, to prevent worsening the COVID-19 pandemic infection rate. However, it is difficult to estimate the degree of congestion within a train without violating passenger privacy. The same issues are true for busses, which is noteworthy because we have previously developed and evaluated a system that can estimate the degree of congestion within a bus while protecting passenger privacy by using Bluetooth Low Energy (BLE) signals. In this paper, we report on our efforts to extend that system to railway use, which were conducted on actual trains in cooperation with Kintetsu Railway Co., Ltd. During this trial, we collected BLE signals and used the data to estimate congestion levels in each car using an ML regression model. The results show that the mean absolute error (MAE) and the mean absolute percentage error (MAPE) could be estimated at accuracy levels of 5.56 and 0.27, respectively.

DOI： 10.1109/di-cps56137.2022.00007

Web of Science

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：LifeTech 2022 - 2022 IEEE 4th Global Conference on Life Sciences and Technologies  

In this paper, we propose a method for estimating psychological states such as hesitation and interest during survey responses based on touch operation behavior (touch log). In addition, we propose two enhanced web survey RUIs (Response User Interfaces), (a) a slidebar type RUI and (b) a magnifier type RUI, to expand the operational differences due to hesitation and interest for more accurate estimation. To evaluate the proposed RUIs, we implemented Augmented Web Survey (AWS) as an extension of LimeSurvey and conducted two experiments. Responses were collected from 8 subjects in the experiment to evaluate (a) a slidebar type RUI and from 21 subjects to evaluate (b) a magnifier type RUI. We measured the degree of hesitation by evaluating the subjective hesitation to the questions on a 5-point scale. For the measurement of the degree of interest, the participants were asked to select their top two favorite images among the six images, and the true value was taken as the top two. As a result of comparing the operation logs of the conventional RUI and the slidebar type RUI, it was confirmed that there was a significant difference depending on the degree of hesitation. In addition, more magnification operations were obtained for the magnifier type RUI than the standard pinch-out operation suggesting that it may help estimate psychological states such as hesitation.

DOI： 10.1109/LifeTech53646.2022.9754964

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/lifetech/lifetech2022.html#NakagawaAN22

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

To promote the face-to-face communication reduced by COVID-19, we proposed and implemented a context-aware Slack chatbot based on Cyber-Physical sensing that helps colleagues meet more often in the same place. Our system periodically collects the user’s internal context through Slack (cyber sensing) and uses small BLE beacons distributed to colleagues and beacon scanners installed in a laboratory to sense physical attendance (physical sensing). In addition, the system notifies the user of recommended actions, such as lunch or coffee break, depending on the context determined by the Cyber-Physical sensors. We deployed the proposed system in a laboratory environment and conducted an initial experiment for six weeks. Experimental results confirmed that our system can encourage serendipitous face-to-face communication during periods when the frequency of attending school and going to work dropped due to COVID-19. It was also found that in an environment such as a laboratory, where a certain level of trust has already been established, the openness of the collected information can further motivate users to participate in the system.

DOI： 10.1007/978-3-030-98438-0_18

Web of Science

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/persuasive/persuasive2022.html#TanakaMNA22

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2022 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events, PerCom Workshops 2022  

In this paper, we propose FedTour, a federated learning-based method for training tourism object recognition models, which utilizes short-distance direct communication between user devices and maximizes the model performance within a limited number of updates. In FedTour, whenever two user devices are within range, they first exchange metadata including the learning degree (e.g., recognition accuracy) of their models, and determine whether it is effective to integrate the peer model by using a regressor trained with various pairs of models with different accuracy to predict the accuracy of the merged model. Once it is deemed effective, the model parameters are exchanged and the model is updated using FedAvg (averaging weights of two models of user devices). By carefully setting the threshold of whether FedAvg is applied or not, model performance is improved within a limited number of model parameter exchanges resulting in lower power consumption of user devices. We conducted a simulation using mobile phone trace data of actual users in a real sightseeing area and evaluated the improvement in accuracy of a CNN model that recognizes 10 objects while limiting the number of model parameter exchanges to only 40. Results show FedTour increased the initial model accuracy by 112&#37;, while the baseline gossip-based method achieved 69&#37;.

DOI： 10.1109/PerComWorkshops53856.2022.9767391

Web of Science

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/percom/percomw2022.html#TomitaTNSY22

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/percom53586.2022.9762376

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：CEUR-WS.org  

In this study, we propose a behavior change support system that encourages people to use stairs instead of elevators. Using the stairs is not only good for one's health, but nowadays it also plays a role in dense avoidance from the perspective of COVID-19. Although many people are waiting for the elevator, we thought that if we knew how many people were on the elevator, more people could change their behavior. However, there are few buildings where the number of people in the elevator is displayed on each floor. So, we have developed a headcount measurement system in the cargo and a visualization system, those can be easily appended to the current elevator. As a method to measure the number of people in an elevator, we propose a method to detect BLE (Bluetooth Low Energy) signals transmitted from the terminals of elevator users who have installed COCOA, an application for confirming contact with the COVID-19 in Japan, and to measure in real time the number of detected BLE signals with a received signal strength (RSSI) exceeding a certain value. We propose a method to measure the number of BLE signals detected in real time. In this paper, we have designed and developed the system, verified the accuracy of the detection of the continuous operation time and the number of passengers, and estimated the behavior pattern and waiting time of elevator users using the system. The evaluation of the transformation of the decision making of the elevator users brought about by this system is out of the scope of this paper and will be discussed in the future.

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/rec/conf/persuasive/2022a

DOI： 10.1109/CPHS56133.2022.00007

Web of Science

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Understanding user emotion is essential for Human-AI Interaction (HAI). Thus far, many approaches have been studied to recognize emotion from signals of various physiological modalities such as cardiac activity and skin conductance. However, little attention has been paid to the fact that physiological signals are influenced by and reflect various factors that have little or no association with emotion. While emotion is a cross-modal factor that triggers responses across multiple physiological modalities, features used in existing approaches also reflect modality-specific factors that affect only a single modality and have little association with emotion. To address this, we propose an approach to extract features that exclusively reflect cross-modal factors from multimodal physiological signals. Our approach introduces a multilayer RNN with two types of layers: multiple Modality-Specific Layers (MSLs) for modeling physiological activity in individual modalities and a single Cross-Modal Layer (CML) for modeling the process by which emotion affects physiological activity. By having all MSLs update their hidden states using the CML hidden states, our RNN causes the CML to learn cross-modal factors. Using real physiological signals, we confirmed that the features extracted by our RNN reflected emotions to a significantly greater extent than the features of existing approaches.

DOI： 10.1007/978-981-99-7019-3_40

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/pricai/pricai2023-1.html#IshikawaKNNIMA22

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Proceedings - IEEE Congress on Cybermatics: 2022 IEEE International Conferences on Internet of Things, iThings 2022, IEEE Green Computing and Communications, GreenCom 2022, IEEE Cyber, Physical and Social Computing, CPSCom 2022 and IEEE Smart Data, SmartData 2022  

The Internet is an effective tool for learners to gain new knowledge. Often, people use search engines (e.g., Google) rather than accessing websites directly. People have their search techniques to find specific information. In particular, people with domain knowledge tend to search more efficiently than novices. By understanding the gap between people with domain knowledge and novices, the novice can understand the path to becoming an expert. Therefore, in this study, we wanted to know what differences exist in search and programming behavior with and without domain knowledge. In this experiment, we asked a group with and without domain knowledge to solve ten programming problems and collected search logs (input knowledge) and compilation logs (output knowledge). Specifically, the first dataset consisted of 13 participants who had taken a university programming class. The second dataset consisted of 20 participants who had not taken a programming class and had no domain knowledge. We examined differences in search and compilation behavior based on participants' domain knowledge from this data. Since we observed a difference between each group when referring to the correlation coefficient, we performed a binary classification of novice and experienced participants using Random Forest, and achieved an average precision of 0.95, indicating that there were different trends in behavior with and without domain knowledge.

DOI： 10.1109/iThings-GreenCom-CPSCom-SmartData-Cybermatics55523.2022.00034

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/ithings/ithings2022.html#WatanabeMNAI22

出版者・発行元：CEUR Workshop Proceedings  

In this study, we propose a behavior change support system that encourages people to use stairs instead of elevators. Using the stairs is not only good for one's health, but nowadays it also plays a role in dense avoidance from the perspective of COVID-19. Although many people are waiting for the elevator, we thought that if we knew how many people were on the elevator, more people could change their behavior. However, there are few buildings where the number of people in the elevator is displayed on each floor. So, we have developed a headcount measurement system in the cargo and a visualization system, those can be easily appended to the current elevator. As a method to measure the number of people in an elevator, we propose a method to detect BLE (Bluetooth Low Energy) signals transmitted from the terminals of elevator users who have installed COCOA, an application for confirming contact with the COVID-19 in Japan, and to measure in real time the number of detected BLE signals with a received signal strength (RSSI) exceeding a certain value. We propose a method to measure the number of BLE signals detected in real time. In this paper, we have designed and developed the system, verified the accuracy of the detection of the continuous operation time and the number of passengers, and estimated the behavior pattern and waiting time of elevator users using the system. The evaluation of the transformation of the decision making of the elevator users brought about by this system is out of the scope of this paper and will be discussed in the future.

Scopus

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：CEUR Workshop Proceedings  

Due to the pandemic of COVID-19, product sampling at supermarkets has been restricted, and the disposal of unsold vegetables and other perishable foods has become a problem. To explore alternatives to product sampling, this study focuses on the “aroma” generated during product sampling and explores the effect of aroma nudges on purchasing behavior. Specifically, we focused on the scenario of promoting the purchase of Yamato-maru eggplant, a traditional vegetable of Nara Prefecture in Japan. We conducted an experiment in a supermarket for two months, from June to July of 2021, the season of the Yamato-maru eggplant. We compared the number of visits to the sales booth, the time spent in the booth, and the sales volume under four conditions: (1) no presentation, (2) presentation of paper media, (3) presentation of the paper and video media, and (4) presentation of a paper, video, and olfactory media. The experiment results suggest that the inclusion of aroma nudges has a positive potential to attract consumer interest and positively influence their purchasing decisions.

Scopus

researchmap

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.23919/ICMU50196.2021.9638884

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3460418.3479284

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PerComWorkshops51409.2021.9431002

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3427477.3429275

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3384419.3430595

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/SMARTCOMP50058.2020.00048

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.3390/smartcities3020013

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.18494/SAM.2020.2615

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.3390/s19204600

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/GCCE46687.2019.9015533

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PERCOM.2019.8767385

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PERCOMW.2019.8730861

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PERCOMW.2019.8730693

記述言語：英語   掲載種別：研究論文（学術雑誌）  

地震などの突発的災害発生時において，避難所への速やかな避難は重要な防災・減災対策の1つである．現状，災害時の避難誘導は，現在地から最寄りの避難所へと誘導するものが一般的である．しかしながら，災害発生時に，大多数の人々が特定エリアに偏って集中するような状況下においては，単に最寄りの避難所に誘導するだけでは避難所の収容数の超過を招き，別の避難所への移動を余儀なくされる．結果として，被災者は空きのある避難所へたどり着くまでたらい回しにされ，避難が大幅に遅れてしまう問題が生じる．本論文では，この問題を解決するために，多くの人々が活動している地域を対象として，被災者全体の避難時間を削減することを目的とする避難所決定手法を提案する．具体的には，被災時における被災者の位置情報と各避難所の収容可能数に基づいて，被災者全体の避難時間が短くなるように避難所を決定する．提案手法の有効性を評価するために，京都の祇園祭を想定したシミュレーション環境を構築し，一般的な避難所決定手法である最短経路選択およびランダム選択を適用した場合と提案手法を適用した場合の避難時間を比較した．その結果，提案手法を用いた場合に，最短経路選択と比較して平均避難時間を35.9&#37;削減できることを明らかにした．また，提案手法に従う人が全体の20&#37;であっても，平均避難時間を8.1&#37;削減できることを確認した．この結果から，たとえ一部の人であっても，提案手法に従って避難することによって，全体の避難時間を短縮でき，防災・減災に寄与できることが分かった．
The quick and appropriate evacuation of the disaster victims is one of great concern in the disaster situation. Under the disaster situations, it is general that the victims evacuate from the present location to the nearest evacuation center depending on evacuation guidance. However, in case of the situation that many victims are at biased distribution to specific areas, the guidance to the closest evacuation center concentrates many victims to particular evacuation. As a result, the victims who could not enter the closest evacuation center need to evacuate to other evacuation centers, until accepted. Therefore, the problem occurs that the evacuation of victims is delayed significantly. In this paper, to solve this problem, we propose an evacuation center determination method aimed at reducing the evacuation time of the whole victims. Specifically, the proposed method determines an evacuation center so that the evacuation time of all victims becomes short based on the location information of victims and the capacity of evacuation centers. To evaluate the effectiveness of the proposed method, we compared our proposed method with the shortest route selection method and random selection method, by using the simulation study assuming the situation of the Gion festival in Kyoto. As a result, when using the proposed method, we showed that the average evacuation time can be reduced to 35.9&#37; compared with the shortest route selection method. Also, we confirmed that the average evacuation time can be reduced to 8.1&#37; even if victims who follow the proposed method is 20&#37; of the total. From these results, even if victims who follow the proposed method is a part of the total, we showed that our method can reduce the overall evacuation time, and contribute to disaster prevention.

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/IWQoS.2018.8624127

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/SEC.2018.00022

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/PERCOMW.2018.8480395

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3267305.3267504

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3267305.3267526

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/3267305.3267624

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/COMPSAC.2018.10307

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1155/2017/6308302

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

A System for Collecting and Curating Sightseeing Information toward Satisfactory Tour Plan Creation

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/ICMU.2016.7742087

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/ICDCSW.2016.37

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/2968219.2968268

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/2968219.2971399

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

Smart Experiment Platform for Wearable Computing

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1109/ICDCS.2016.101

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1145/2645884.2645888

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2044929

記述言語：日本語  

新型コロナウイルス感染拡大により，実店舗での試食販売の実施が制限され，売れ残った野菜などの生鮮食品の廃棄が問題となっている．本研究では，試食販売の代替案を模索するべく，試食販売時に発生する「香り」に着目し，香りのナッジが購買行動に及ぼす影響を調査する．具体的には，奈良県の伝統野菜である大和丸なすの購買促進を題材として取り上げ，大和丸なすの旬である2021年の6月から7月の2ヶ月間，実店舗での実証実験を行なった．そして，（1）提示なし，（2）紙媒体の提示，（3）紙・動画媒体の提示，(4）紙・動画・香り媒体の提示という４つの条件下における，販売ブースへの訪問回数，滞在時間，売上高を比較した．実験の結果，香りナッジを用いた介入によって消費者の興味を引き，購買の意思決定に正の影響を与える可能性が示唆された．
Due to the pandemic of COVID-19, tasting sales at supermarkets have been restricted, and the disposal of unsold vegetables and other perishable foods has become a problem.To explore alternatives to tasting sales, this study focuses on the "aroma" generated during tasting sales and explores the effect of aroma nudges on purchasing behavior.Specifically, we focused on the scenario of promoting the purchase of Yamato-maru eggplant, a traditional vegetable of Nara Prefecture in Japan. We conducted an experiment in a supermarket for two months, from June to July of 2021, the season of the Yamato-maru eggplant. We compared the number of visits to the sales booth, the time spent in the booth, and the sales amount under four conditions: (1) no presentation, (2) presentation of paper media, (3) presentation of the paper and video media, and (4) presentation of a paper, video, and olfactory display media. The results of the experiment confirmed that sales increased by an average of 1.83 times from the previous year due to the intervention using aroma nudges. Through the demonstration experiment, it was suggested that interventions using aroma nudges have a positive potential to attract consumer interest and positively influence their purchasing decisions.

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

DOI： 10.11517/jsaisigtwo.2020.sai-039_10

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：英語  

DOI： 10.1016/j.margeo.2016.08.006

記述言語：日本語  

記述言語：日本語  

記述言語：英語  

記述言語：英語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

大規模災害時に被災状況を迅速に把握し,災害救助活動を支援するための情報通信システムの構築が求められている.しかし,そのシステムを社会で実用化するためには,導入時だけでなく運用時に要する経費や,災害用のシステムを平常時から運用するための仕組みなど様々な課題があり,普及が進んでいない.特に,研究機関の研究成果と耐災害システムのユーザである自治体等の要求をマッチングさせること,非常時に適切に稼働することが求められる.本研究では,情報通信機構(NICT)で開発された耐災害メッシュネットワークNerveNetを教育活動に導入し,その教育活動を通して災害時情報通信システムの社会実装を図ることを目的とする.多くの高等専門学校では,地域密着型教育活動を推進するとともに,実践的教育活動の中で実用に供するシステム構築を行っている.特に,函館高専では,専攻科で実施されている創造実験Project-based Learning(PBL)で,地域社会からの要請事項を実験テーマとして取り上げ,課題解決型教育活動の中で課題の解決を図り社会に還元している.その取り組みの中で耐災害システムの構築を実験テーマとして設定し,函館市やその他周辺自治体の要求を調査し,NerveNetを活用したテストベッドを構築した.テストベッドでは,安否情報管理やファイル管理機能をWebサーバによって提供し,その通信ネットワークにNerveNetを使用した.また,サーバ型カメラをシステムに導入して,映像による遠隔モニタリング機能を実装した.このテストベッドによって,自治体が求める要求を具体化して,有効に機能するシステム構築に役立てる.また,その教育活動による技術修得によって,災害時にシステム運用を支援する.

記述言語：日本語  

大規模災害時に迅速に被災状況を把握し,自治体等の災害救助活動を支援する災害時情報通信システムの研究が多くの機関で実施されてきた.しかし,そのシステムを社会で実用化する社会実装には様々な課題があり,自然災害が多発している日本においても普及しているとは言い難い.本稿では,まず災害時通信システムとして,情報通信機構(NICT)で開発された耐災害メッシュネットワークNerveNetについて記述する.そのNerveNetを被災状況把握システムに活用するにあたり,自治体の要求を調査し,その要求を実現する災害時情報管理システムについて検討する.これらのシステム構築と社会実装に関して,地域密着型教育活動を推進している高等専門学校の教育活動とNICT及び地方自治体との連携において実施し,その教育活動によって応用システムの試作と災害時通信システムの運用に関する取り組みについて報告する.

記述言語：英語  

DOI： 10.1016/j.epsl.2014.03.043

記述言語：日本語  

センサネットワーク技術は,多様な分野への応用が可能な技術であり,これまで数多くのセンサネットワークシステムが開発されてきた.しかし,現在使用されているセンサネットワークの多くが特定のアプリケーションに特化した専用のシステムとなっている.そこで我々は,異なる用途のセンサネットワークやその他の機器が連携することができるミドルウェアアーキテクチャとしてCMSAS(Common Middleware in Sensors,Actuators and Storagors)の検討を行っている.CMSASでは,システムを構成する全ての機器に共通のミドルウェアを導入し,共通のミドルウェアのルールに基づき機器の管理を統一化する.これによって,機器間の仮想的な相互の連携を実現することを目指す.本稿では,検討の第一段階として,産業技術総合研究所(AIST)で開発されたRTミドルウェアを利用したCMSASの概念について紹介し,センサネットワークを用いた加速度データ収集システムへのCMSASの実装について述べる.

記述言語：日本語  

地震等大規模災害時に橋梁やトンネル,建物などの構造物に生じる加速度や歪を計測し,その構造物の状況を把握する構造物ヘルスモニタリングの研究が様々な研究機関で進められている.本研究は,建物に設置された加速度センサによって地震加速度を検出し,そのデータを無線センサネットワークで収集して,構造物ヘルスモニタリングシステムに情報を提供するデータ収集システムについて検討する.本稿では,地震時に観測された加速度信号を分析して帯域について考察し,センサネットワークで加速度データを収集するための通信制御方式について検討する.その通信制御方式の有効性を確認するため,無線センサネットワークのテストベッドを構築し,その省電力特性とデータ伝送特性について評価を行った.ここでは,地震加速度信号の分析と無線センサネットワークテストベッドによる評価実験について記す.

記述言語：英語  

DOI： 10.1016/j.sedgeo.2013.06.004

記述言語：日本語  

無線センサネットワーク(WSNs)を利用した地震加速度モニタリングに関する検討が行われおり,地震加速度の収集システムにおいては,数10～100Hzのサンプリングレートが要求される.これまでに著者らは,地震加速度の特性を調査したうえ,WSNs, Wireless Distribution System,データベースサーバから構成される加速度モニタリングシステムのテストベッドを構築した.テストベッドによる基礎実験から,WSNsではセンサノードからシンクノードへのシングルホップ通信において,サンプリングレート100Hzで生成されるデータを連続的に伝送することが可能であることを示した.本稿は,テストベッドを用いて加速度データをマルチホップ伝送し,データベースへ格納する実験結果について報告する.結果として,データの収集過程に損失が発生し,その原因はWSNsのマルチホップによる影響だけではなく,データベースへの格納時にも生じることを確認した.

記述言語：日本語  

近年、ユビキタス社会実現の基盤となる技術として無線センサネットワークが注目され、災害時の情報収集システムに応用することが期待されている。本研究では、加速度センサで地震加速度を計測し、無線センサネットワークを用いて加速度データを収集する技術を検討するとともに、そのデータを管理するモニタリングシステムの開発を行っている。本発表では、無線センサネットワークで収集され、大量に送られてくる加速度データの管理と、リアルタイムな情報提供を実現するWebサーバによるデータの提供方法について提案する。

記述言語：日本語  

無線センサネットワークの応用例として,地震加速度をモニタリングする研究が注目されている.無線センサネットワークを用いた地震加速度モニタリングでは,高サンプリングレートによる連続的な地震加速度データの収集,駆動電力の省電力化,モニタリング領域の拡張性という3つの要件を考慮せねばならない.しかし,対象とする領域が広域である場合,無線センサネットワークによる多段のマルチホップではスループットの低下や消費電力の増大により要件を満たすことが難しい.そこで,本研究では無線センサネットワーク領域を小クラスタに分け,無線LAN技術によって無線領域を拡張し,インターネットを介して地震加速度データを収集するモニタリングシステムを提案した.本稿では3層のネットワーク技術から成る地震加速度モニタリングシステムの設計を示したうえ,無線センサネットワーク領域に適した伝送方式を調査するべく加速度データ収集の基礎実験を実施し,伝送レートおよび伝送効率の特性を評価した.

記述言語：日本語  

地震加速度のモニタリングをセンサネットワークで行う場合、限られた通信帯域で連続してデータ収集するとともに省電力動作を実現することが求められる。しかし、これまで研究されたセンサネットワークの通信制御方式では、この要求を満たすことが困難である。そこで本研究では、地震による構造物の振動特性に注目し、サンプリング周波数を最適化して通信制御することによって消費電力を軽減する方式を検討している。本発表では、サンプリング周波数に応じて通信制御した場合の省電力効果について示す。

記述言語：英語  

DOI： 10.1177/0022487112447110

種別：大会・シンポジウム等